The field of the invention is nuclear magnetic resonance (NMR) systems and methods and, more particularly, systems for calibrating the RF transmit section of an NMR imaging system.
NMR imaging employs a combination of pulsed magnetic field gradients and pulsed radio frequency fields to obtain NMR imaging information from nuclear spins situated in selected regions of a sample to be imaged. The sample is typically positioned in a static magnetic field effective to polarize nuclear spins having a net magnetic moment so that a greater number of spins align with the polarizing magnetic field and add to produce a net magnetization. The individual polarized nuclear spins, and hence, the net magnetization M, resonate, or precess about an axis of the polarizing magnetic field at a frequency equal to a gyromagnetic constant times the magnitude of the magnetic field. This relationship is known as the Larmor relationship. A gyromagnetic constant exists for each NMR isotope. For a hydrogen nucleus (the most abundant atomic nucleus in a living body), the gyromagnetic constant is about 42.58 MHz/Tesla. With a magnetic field of about 1.5 Tesla, for example, the resulting resonant frequency for hydrogen nuclei predicted by the Larmor relationship is about 63.9 MHz.
In order to obtain a detectable NMR signal, the net magnetization of the nuclear spins is rotated away from coincidence with the axis of the polarizing magnetic field. Rotation is performed using a radio frequency excitation field of the same frequency as that determined by the Larmor relationship. The angle through which the net magnetization is rotated, or "flipped", is related to the field strength of the radio frequency excitation signal and to its duration. At the end of the radio frequency excitation pulse, the nuclei, in relaxing to their normal spin conditions, generate a decaying signal at the same radio frequency as that used for excitation. This NMR signal is picked up by a receive coil, amplified and processed by the NMR system.
As noted above, the angle through which the net magnetization is rotated depends on the radio frequency field strength and duration. NMR imaging generally requires that the net nuclear magnetization be rotated by some specified angle. Rotation to angles varying from these specific angles can produce a variety of problems in the reconstructed image. A full discussion of such problems and one solution therefor is given in U.S. Pat. No. 4,443,760, the disclosure of which is incorporated herein by reference.
Besides adding ghost artifacts to the received signal, errors in radio frequency field strength and pulse duration substantially reduce the amplitude of the received NMR signal. Since the received radio frequency NMR signal is small at best, reduction thereof, with a concomitant degradation in signal-to-noise ratio, is not desirable. Errors in radio frequency field strength can also affect image contrast in low flip angle gradient recalled echo sequences. The flip angle determines the relative proportions of T1 and T2 contrast. Errors in the flip angle could even produce the opposite of the intended contrast, yielding a T2-weighted image when a T1-weighted image was desired.
The combination of radio frequency field strength and pulse duration required to produce a particular rotation of net magnetization varies from object to object being imaged. In general, the more massive the body to be imaged, the higher the field strength and/or pulse duration. Also, the required field strength and duration varies with the type of material through which the exciting radio frequency pulse must travel to excite the material being imaged. When the body being imaged is a portion of a human anatomy, for example, the excitation by the radio frequency field varies with patient weight, the portion of the body being imaged, and the proportion of body fat, among other things.
Before the commencement of each NMR scan, it is common practice to adjust the frequency of the RF transmitter and receiver to insure that the RF excitation field is at the optimal Larmor frequency. Such a procedure is disclosed in U.S. Pat. No. 4,806,866, which is entitled "Automatic RF Frequency Adjustment For Magnetic Resonance Scanner", and which describes a calibration sequence that automatically determines the best RF transmitter and receiver frequency at the beginning of each NMR scan.
Similarly, before the commencement of each NMR scan, it is common practice to adjust the strength of the transmitted RF excitation field and the gain of the RF receiver so that accurate 90.degree. and 180.degree. flip angles are produced by the RF excitation field pulses. Methods for making this adjustment automatically are also well known and used in all commercially available MRI systems.
While the above described techniques insure that the RF excitation pulses have the optimal frequency, strength and duration to evoke the desired NMR signal, this does not necessarily mean that the expected RF excitation field will be produced uniformly throughout the region of interest, or that the resulting NMR signals will be received uniformly from all locations in the region of interest. Indeed, most transmit coils are loaded by the subject being studied, and the RF fields produced are not homogeneous. This is particularly true of so-called local coils which are relatively small coils that are designed to image specific, relatively small regions of human anatomy.
Prior RF power calibration methods perform a prescan sequence in which one or more NMR measurements are performed to determine the optimal RF power setting. Typically, a projection acquisition is performed in which a spin echo pulse sequence is used, but no phase encoding gradient is applied. The acquired NMR signal is an accumulation of the spin signals produced throughout the field of view of the measurement and these spin signals are weighted evenly regardless of their location. Such uniform sampling of the NMR signal level is appropriate when the prescribed scan uses the imager's whole-body rf coil that is designed to produce a homogeneous transmit and receive field throughout the entire field of view.
When local coils are prescribed, prior RF power calibration methods do not optimize the RF power setting for the small, local region of interest these coils are designed to image. Local coils enhance the level of the NMR signals received from a small region adjacent to the coils, but their response falls off drastically for signals located further from the coil. A calibration method which relies on a projection measurement that evenly weights signals throughout the field of view will not accurately set the rf power for the region of interest, which may or may not be immediately adjacent to the coil.